OBJECTIVE - We suggest a simple, noninvasive method to assess the auto
nomic function in diabetic subjects. The method requires only a monito
ring of heart rate (HR) with subjects in the sitting position. RESEARC
H DESIGN AND METHODS - Sixty diabetic subjects, 44 men and 16 women, b
etween 20-80 years of age, were recruited, chronologically, for this s
tudy. Subjects treated for high blood pressure were not included. Thei
r autonomic function was assessed by the total score of five classical
cardiovascular function tests. In the same subjects and in 44 healthy
subjects, blood pressure and HR were determined from beat to beat by
the Finapres system with subjects in the sitting position. We examined
the randomness of the HR changes by calculating the z statistic of th
e runs test on 1,000 successive HR readings (the z value is low if the
HR changes are random). When the HR changes are random, we consider t
hat the autonomic control of HR is impaired. RESULTS - The z values of
HR changes were significantly lower in diabetic subjects compared wit
h normal subjects (2.98 +/- 0.97 vs. 3.54 +/- 0.97, P < 0.004). In dia
betic subjects, the z value was closely correlated to the total score
of disautonomy (r = - 0. 66, P < 0.0001, after correction for age effe
ct) and to the office systolic blood pressure (r = -0.43, P < 0.001).
CONCLUSIONS - The z value of HR changes might be a marker of the auton
omic function in diabetic subjects.